Build/Update Tests Skill

Evaluate test coverage and create or update tests when gaps or pattern violations are identified.

<whentouse>

When to Use

Invoke when user says:

• "build tests for [file]"
• "update tests for [file]"
• "test [component/service/hook]"
• "add test coverage for [file]"
• "check test coverage"

</whentouse>

<workflow>

Workflow Overview

For detailed phase instructions: [references/execution-phases.md](references/execution-phases.md) </workflow>

<risk_matrix>

Risk Matrix

Code Type Requirements

</risk_matrix>

<test_locations>

Test File Locations

</test_locations>

<pattern_compliance>

Pattern Compliance

Check these pattern documents before writing tests:

• claude-patterns/testing-patterns.md (React component unit tests)
• claude-patterns/vitest-testing-patterns.md (Service/utility unit tests)
• claude-patterns/playwright-best-practices.md (E2E tests)

Priority Order

Fix P0 violations before creating new tests—a test that violates patterns is worse than no test.

For detailed patterns: [references/best-practices.md](references/best-practices.md) </pattern_compliance>

<invocation>

Invocation

Basic:

Use build-update-tests skill for src/components/MyComponent.tsx

With requirements:

Use build-update-tests for src/services/email.service.ts with critical flows from docs/email-requirements.md

</invocation>

<commands>

Commands

# Run unit tests for specific file
npm run test:unit -- [test-file-name]

# Run all unit tests
npm run test:unit

# Run E2E tests
npx playwright test [spec-file]

</commands>

<output>

Output

Produce a comprehensive report using [references/report-template.md](references/report-template.md):

1. Evaluation Phase: Risk assessment, test discovery, pattern compliance, gap analysis
2. Implementation Phase: Actions taken, tests created/updated, coverage achieved
3. Summary: Strengths, warnings, gaps fixed, recommendations

</output>

<approval_gates>

Approval Gates

Note: Test creation is non-destructive. User reviews output report and decides what to keep. </approval_gates>

<safety_rules>

Safety Rules

1. Never delete existing passing tests - only add or update
2. Fix P0 pattern violations before creating new tests - bad tests are worse than no tests
3. Preserve existing test structure - don't reorganize without clear benefit

</safety_rules>

<limitations>

Limitations

• Cannot create E2E tests without understanding full user workflow
• Cannot determine test priority without requirements documentation
• Complex mocking may require code refactoring for testability

</limitations>

<quick_reference>

Quick Reference

# Find existing tests
glob "src/**/__tests__/**/*.test.{ts,tsx}"
glob "playwright/tests/*.spec.ts"

# Check for pattern docs
cat claude-patterns/testing-patterns.md
cat claude-patterns/playwright-best-practices.md

</quick_reference>

<references>

References

• [references/execution-phases.md](references/execution-phases.md) - Detailed 6-phase workflow
• [references/report-template.md](references/report-template.md) - Output report format
• [references/best-practices.md](references/best-practices.md) - Testing patterns and examples

</references>

<version_history>

Version History

• v3.2.0 (2025-01-18): AI optimization updates

- Add blockquote summary after title - Soften directive language per Opus 4.5 guidance

• v3.1.0 (2025-12-28): Added missing sections per skill-authoring-patterns audit

- Added trigger phrases to frontmatter description - Added <approvalgates> section - Added <safetyrules> section - Added <references> section

• v3.0.0 (2025-12-28): Refactored to follow skill-authoring-patterns

- Reduced from 850 to ~180 lines - Created references/: execution-phases.md, report-template.md, best-practices.md - Added XML tags for structure - Converted to imperative writing style

• v2.0.0 (2025-11-15): Combined audit and implementation workflows

• v1.0.0 (2025-10-01): Initial release

</version_history>